Site-specific O-glycans influence lacritin structure and multimerization in tears

This study utilizes mass spectrometry-based glycoproteomics and molecular dynamics simulations to demonstrate that site-specific O-glycans on lacritin modulate its structural flexibility and multimerization by influencing the conformation and solvent accessibility of key crosslinking residues (Lys101 and Lys104), thereby regulating its biological functions in tear fluid.

The Gist

Imagine your tears aren't just salty water; they are a sophisticated, high-tech security system for your eyes. At the heart of this system is a protein called Lacritin. Think of Lacritin as a "Swiss Army Knife" for your eye: it helps produce tears, fights off bacteria, and repairs damaged skin on the eye's surface.

However, for decades, scientists have been puzzled by one thing: Lacritin is covered in a thick, sugary coat (called O-glycans) that makes up more than half of its weight. It's like a superhero wearing a heavy, fuzzy costume. We knew the costume was there, but we didn't know if it was just decoration or if it actually changed how the superhero fought.

This paper solves that mystery by looking at two different "modes" of Lacritin: the Single Agent (monomer) and the Team (multimer).

The Problem: The "Team" Can't Work

Lacritin works best when it's a single agent. When it's alone, it can lock onto a specific receptor (like a key in a lock) to tell your eye to produce more tears and heal itself.

But sometimes, Lacritin gets cross-linked by an enzyme (think of it as a molecular glue) to form a "Team" or a chain of proteins. The problem? Once they form a Team, they get too bulky and clumsy to fit into that keyhole. They can't do their job anymore.

The Discovery: The Sugar Coat is the Switch

The researchers asked a simple question: Does the sugary coat change depending on whether Lacritin is a Single Agent or a Team?

Using a high-tech microscope (Mass Spectrometry) and a clever filtering method (like a sieve that separates big rocks from small pebbles), they separated the Single Agents from the Teams and analyzed their sugar coats.

They found a major difference:

• The Single Agents had a very specific, heavy sugar coating near their "hands" (the parts that grab onto the eye's receptors).
• The Teams had a much lighter, different sugar coating in those same spots.

The Simulation: A Molecular Dance

To understand why this matters, the scientists used a computer to build 3D models of these proteins and ran a "dance simulation" (Molecular Dynamics) to see how they moved.

Here is what they discovered using creative analogies:

1. The "Sticky" Sugar: On the Single Agents, the sugar molecules acted like Velcro. They stuck to the protein's backbone, holding the protein in a tight, rigid shape. This rigidity kept the "hands" (the cross-linking spots) hidden or blocked, preventing the protein from accidentally gluing itself into a Team.
2. The "Free" Sugar: On the Teams, the sugar coating was different. It didn't stick to the protein as much. This made the protein more flexible and floppy, like a wet noodle compared to a stiff stick.
3. The "Open Door" Effect: Because the Team version was floppier, its "hands" (specifically two spots called Lys101 and Lys104) were exposed and waving in the air, ready to grab onto other proteins and form a chain. The Single Agent's "hands" were tucked away and protected by the sugar coat.

The Big Picture

Think of the sugar coat as a traffic controller.

• When the sugar coat is thick and specific (Single Agent), it acts like a red light, telling the protein: "Stay put, don't glue yourself to others, go do your job as a single unit."
• When the sugar coat is different (Team), it acts like a green light, allowing the protein to become flexible, expose its sticky hands, and form a chain.

Why Does This Matter?

This is huge news for people with Dry Eye Disease.

• If the sugar coat gets messed up (due to disease or aging), the "traffic controller" might fail.
• The Single Agents might accidentally turn into Teams, losing their ability to heal the eye.
• Or, the Teams might not form when they are needed for protection.

In short: This paper reveals that the sugary coat isn't just decoration; it's a structural switch that decides whether Lacritin works as a healing hero or gets stuck in a useless chain. By understanding this, scientists can design better drugs (like the peptide "Lacripep" mentioned in the paper) that mimic the perfect sugar-coated shape to treat dry eyes more effectively.

Technical Summary

1. Problem Statement

Lacritin is a critical glycoprotein in tear fluid (~23 kDa) involved in tear secretion, immune response, and antimicrobial activity. Its function is tightly regulated by post-translational modifications, specifically O-glycosylation, which constitutes over 50% of its molecular weight. Despite its importance as a potential biomarker and therapeutic target for Dry Eye Disease (DED), the functional role of these O-glycans remains largely unknown due to:

• Analytical Complexity: The high heterogeneity of O-glycans and the low abundance of lacritin in complex biological fluids (tears) have hindered site-specific characterization.
• Structural Blind Spots: Traditional structural biology methods (X-ray crystallography, Cryo-EM) struggle with the conformational flexibility of glycosylated proteins.
• Regulatory Mechanism Gap: While it is known that lacritin multimerization (via transglutaminase TGM2 crosslinking at Lys101 and Lys104) inhibits its binding to syndecan-1 (SDC1) and downstream signaling, the specific influence of O-glycans on this multimerization process and the resulting structural topology has not been elucidated.

2. Methodology

The authors employed a multi-disciplinary workflow combining glycoproteomics and molecular dynamics (MD) simulations to bridge the gap between sequence data and structural function.

• Sample Preparation (GlycoFASP):
  • Tear fluid was collected from healthy donors.
  • A filter-aided sample preparation (FASP) strategy using molecular weight cutoff (MWCO) filters was developed to physically separate monomers (<30 kDa) from **multimers** (>30 kDa).
  • The separated fractions were treated with SmE mucinase (an O-glycoprotease) to release O-glycopeptides, followed by trypsin digestion.
• Mass Spectrometry (MS) Analysis:
  • LC-MS/MS was performed using an Orbitrap Eclipse Tribrid with HCD-pd-ETD (Higher-Energy Collisional Dissociation coupled with Electron Transfer Dissociation) fragmentation to ensure confident identification of O-glycopeptides.
  • Data was processed using Byonic for database searching and manual validation of spectra.
  • Label-Free Quantitation (LFQ) using Area Under the Curve (AUC) intensities was used to compare glycan abundances between monomer and multimer fractions.
• Computational Modeling:
  • AlphaFold 3.0 and CHARMM-GUI were used to generate initial 3D structures of the lacritin C-terminal domain (residues 79–137) with site-specific glycosylation based on the most abundant MS-identified glycoforms.
  • Molecular Dynamics (MD) Simulations: 500 ns production runs were executed using GROMACS on three constructs:
    1. Non-glycosylated (unmodified).
    2. "Monomer" glycoform (based on monomer MS data).
    3. "Multimer" glycoform (based on multimer MS data).
  • Analysis Metrics: Root Mean Square Deviation (RMSD) for backbone flexibility and Solvent-Accessible Surface Area (SASA) for crosslinking residues (Lys101, Lys104).

3. Key Contributions

• First Site-Specific Comparison: This study provides the first in-depth glycoproteomic comparison of endogenous lacritin monomers versus multimers in native tear fluid.
• Novel Workflow: The development of a GlycoFASP protocol specifically optimized for separating and characterizing lacritin oligomers from complex tear fluid.
• Integration of MS and MD: The paper demonstrates a scalable framework for integrating site-specific glycoproteomics data with accessible MD simulation tools (CHARMM-GUI/GROMACS) to predict structural impacts of glycosylation without requiring specialized supercomputing resources or manual model building.

4. Key Results

A. Distinct Glycosylation Profiles

• Separation Validation: Western blot and MS confirmed the successful separation of monomers and multimers. Notably, peptides containing Lys104 (a crosslinking site) were only detected in the monomer fraction, confirming that crosslinking prevents tryptic cleavage or detection in the multimer fraction.
• Site-Specific Differences: Significant differences in O-glycan heterogeneity were observed at Ser86, Ser91, and Thr95:
  • Ser86: Multimers showed higher abundance of complex Core 2 and disialylated Core 1 structures compared to monomers.
  • Ser91: Monomers were predominantly modified with monosialylated Core 1 (83.95%), whereas multimers were largely unoccupied (42.77%) or had only Tn antigen (GalNAc).
  • Thr95: Both populations were primarily Tn antigen-modified, but multimers had slightly higher occupancy.

B. Structural Dynamics and Intra-Glycan-Protein Interactions (GPIs)

• Intra-GPIs: MD simulations revealed that glycans form specific polar contacts (hydrogen bonds, pi-stacking) with the protein backbone.
  • Monomer: Glycans at Ser86 and Ser91 interact with Asn82, Lys104, and Ile92, stabilizing the local structure.
  • Multimer: The absence of glycans at Ser91 and different occupancy at Ser86 led to altered interactions (e.g., Leu81 and Lys90 interactions), changing the spatial arrangement of the C-terminal helices.
• Backbone Flexibility: The unmodified (non-glycosylated) construct exhibited significantly higher flexibility (RMSD 1.4 nm) compared to glycosylated forms (0.6 nm). This suggests O-glycans impart structural rigidity to the C-terminal domain.

C. Solvent Accessibility of Crosslinking Residues

• SASA Analysis: The study quantified the Solvent-Accessible Surface Area (SASA) of the critical crosslinking residues Lys101 and Lys104.
• Findings:
  • The multimer-associated glycoform exhibited significantly higher SASA for Lys101 and Lys104 compared to both the monomer glycoform and the unmodified structure.
  • In the monomer glycoform, bulky sialylated Core 1 glycans were positioned close to Lys101/104, potentially sterically hindering access.
  • Conclusion: Proximal O-glycosylation patterns dictate the solvent exposure of crosslinking sites, thereby modulating the propensity for TGM2-mediated multimerization.

5. Significance

• Mechanistic Insight: The study establishes a direct link between site-specific O-glycosylation and the structural topology of lacritin. It proposes a regulatory model where specific glycan patterns at Ser86/91/Thr95 control the accessibility of Lys101/104, thereby acting as a "switch" for multimerization and subsequent loss of SDC1-binding activity.
• Therapeutic Implications: Understanding how glycans regulate lacritin function is crucial for developing lacritin-based therapeutics (e.g., Lacripep) for Dry Eye Disease. It suggests that glycosylation status could be a critical variable in drug efficacy or stability.
• Diagnostic Potential: The distinct glycosylation signatures between monomers and multimers offer new avenues for biomarker discovery in ocular pathologies.
• Methodological Advancement: The paper provides a reproducible, accessible pipeline for studying other O-glycosylated proteins, moving beyond static structural predictions (like AlphaFold) to dynamic, glycan-informed structural biology.

In summary, this work resolves a decades-long gap in understanding lacritin biology by demonstrating that O-glycans are not merely passive decorations but active structural determinants that regulate protein flexibility, multimerization, and biological function at the ocular surface.